Techno-Economic and Environmental Assessment of a Hybrid Photovoltaic–Diesel–Grid System for University Facilities

Abstract

This study presents a techno-economic and environmental assessment of a photovoltaic–diesel–grid hybrid renewable energy system (SHER) applied to a university campus, with the aim of reducing monetary costs by implementing a methodology to mitigate energy consumption during peak hours, controlling the output of the diesel generator, and determining greenhouse gas emissions. Hourly load profiles are incorporated using billing data, local solar resource data, and grid connection rate schedules. The HOMER Pro v3.14.2 software is used to simulate and identify an off-grid scenario during peak hours, sizing the photovoltaic system at 30%, 50%, 70%, and 100% to compare the investment cost of the SHER. System performance is evaluated using key indicators, including net present cost ($6.96 million), levelized cost of energy (LCOE, $0.707/kWh), and CO₂ emissions (101,311 kg/yr.), among others. The results indicate that photovoltaic generation can cover approximately 80% of annual electricity demand, while the diesel generator operates only during critical periods, contributing to reduced operating costs and emissions. The optimal configuration has a lower LCOE than conventional supply, a renewable fraction of close to 80%, and an investment payback period of approximately five years, demonstrating the technical, economic, and environmental viability of the proposed system.

1. Introduction

The global transition towards sustainable, low-carbon energy systems has intensified interest in hybrid renewable energy systems (HRESs), particularly in applications characterized by high electricity demand, complex tariff schemes, and strict reliability requirements [1]. Educational institutions are a strategic sector for the implementation of these systems due to their relatively predictable load profiles, predominantly daytime operation, and the need to reduce operating costs without compromising sustainability objectives [2].

In several countries, medium-voltage users are subject to time-of-use tariff schemes that include base, intermediate, and peak periods, as well as charges associated with peak demand, capacity, and use of the electrical infrastructure. For university campuses, these components can represent a significant proportion of annual billing, especially when peak demand coincides with high-cost tariff periods, limiting the efficiency of strategies based exclusively on grid supply [3,4].

Photovoltaic generation reduces energy consumption during periods of high irradiance; however, systems based on a single renewable technology have limitations associated with intermittency and limited control over peak demand. In this context, HRESs that integrate renewable sources with dispatchable technologies, such as diesel generators, emerge as a viable alternative to improve supply reliability and the techno-economic and environmental performance of the energy system [5]. Despite this, studies focused on educational campuses that explicitly incorporate complex time-of-use tariff schemes remain limited [6,7,8]. The literature shows there are a high percentage of hybrid systems used in educational areas, compared to industrial applications and rural communities [9].

Current buildings represent one of the main sources of energy consumption and greenhouse gas emissions due to the operation of heating, ventilation, air conditioning, and other associated equipment [10]. Although various studies have sought to reduce the energy consumption of buildings through renovations that incorporate advanced technologies, such as improved insulation, double-skin facades, and phase-change materials, the integration of renewable energy generation systems is a key alternative to offset the total energy demand of these buildings [11,12].

Global CO₂ emissions associated with the energy sector increased by 0.8% in 2024, reaching a historic high of 37.8 Gt of CO₂. This increase contributed to the atmospheric concentration of CO₂ rising to 422.5 ppm, approximately 3 ppm higher than in 2023 and 50% above pre-industrial levels, thus consolidating an upward trend in the accumulation of greenhouse gases. During 2024, emissions from fossil fuel combustion increased by around 1%, equivalent to 357 Mt of CO₂, while emissions from industrial processes showed a decrease of 2.3%, corresponding to 62 Mt of CO₂. It should be noted that emissions growth remained below global GDP growth (+3.2%), restoring the long-term trend toward decoupling economic growth from emissions growth, a phenomenon that had been temporarily interrupted in 2021 [13].

The study by Abdelsattar presents a grid-connected hybrid microgrid model that integrates photovoltaic, wind, and diesel generation, optimized using HOMER Pro [14]. Its main contribution lies in the joint evaluation of economic and environmental variables, demonstrating a significant reduction in operating costs and CO₂ emissions. It also demonstrates that the diesel generator can operate as a strategic backup in systems with high renewable penetration. However, the study has limitations regarding stability analysis, protection, and advanced control strategies, which opens up opportunities for future research aimed at the comprehensive optimization of hybrid systems under specific tariff conditions [14].

A study has shown that the penetration of photovoltaic generation in PV–diesel hybrid systems is limited primarily by the operational constraints of the thermal system, rather than by the availability of solar resources. By using passive strategies, such as geographic dispersion and optimization of diesel dispatch, renewable penetration can be increased to up to 30% without the need for storage systems. These results highlight the importance of considering operational stability in the planning of hybrid systems [15].

The study by Mulenga evaluates the technical and economic feasibility of hybrid photovoltaic-diesel systems in rural areas without access to the power grid, using the HOMER Pro tool to optimize energy configurations based on levelized cost of energy (LCOE) and lifecycle cost criteria [16]. The results show that stand-alone diesel systems have high operating costs and low economic sustainability, while the integration of photovoltaic generation significantly reduces fuel consumption and improves system efficiency. Furthermore, the study finds that 100% renewable systems have the lowest LCOE, albeit with higher initial investment costs. However, the study is limited to off-grid systems, without considering interaction with the power grid or tariff structures, which presents an opportunity for future research on grid-connected hybrid systems [16].

Table 1. Consolidated comparative table—hybrid energy systems (diesel-based, 2023–2026).

Ref.	Year	System Type	Technical Parameters	Economic Parameters	Environmental Parameters	Methodological Approach
[17]	2023	Stand-alone microgrid (PV–Wind–Diesel–Battery)	Irradiance, wind, temperature, SOC, fuel consumption	NPC, LCOE, CAPEX, OPEX	Renewable fraction	HOMER + sensitivity analysis
[18]	2023	Hybrid system (PV–Wind–Diesel–Battery, grid and off-grid)	Load profile, generation mix, sizing	IRR, VPN, payback	Emission reduction	HOMER simulation
[19]	2023	Hybrid PV–Diesel–Battery industrial system (Algeria)	Power generation, PV penetration PV output, diesel share, load	NPC, cost optimization. LCOE	Emission reduction (~60%), fuel savings	HOMER + sensitivity
[20]	2023	Hybrid system tunnels/infrastructure	Annual production, configuration	LCOE ≈ 0.17 $/kWh	GHG reduction	HOMER + scenarios
[21]	2023	Floating PV + Diesel system	PV capacity, load, diesel operation	Fuel savings	CO2 reduction	System design + feasibility
[22]	2023	Review of hybrid microgrids (diesel backup)	Configurations, storage, sizing	LCOE, NPC	Emission reduction	Systematic review
[5]	2024	Hybrid microgrid EMS (PV–Wind–Diesel–Battery)	Energy flow, SOC, control	Operational optimization	Emission reduction	MATLAB/Simulink
[23]	2024	Hybrid minigrid (PV–Diesel–Grid)	Load, generation, reliability	NPC, LCOE	Emission reduction	HOMER optimization
[24]	2024	Hybrid PV–Diesel system (remote Algeria)	Solar radiation, load	LCOE ≈ 0.172 $/kWh	CO2 reduction	HOMER
[25]	2024	Hybrid off-grid PV–Diesel–Battery (Indonesia)	Load, PV capacity, diesel sizing	COE reduction (~15.7%), NPC	Emission reduction	HOMER
[8]	2024	Campus microgrid (renewables + diesel backup)	PV, wind, geothermal	Installation cost	Low emissions	Systematic review
[26]	2024	Hybrid system with hydrogen + diesel	Energy balance, H2 integration	Economic optimization	Emission reduction	Advanced optimization
[27]	2024	Hybrid microgrid dispatch (diesel + RES + storage)	Dispatch variables, DER	Cost minimization	Emission reduction	Linear programming
[28]	2024	HRES review (PV–Diesel–Battery systems)	Configurations, reliability	Lifecycle cost	Environmental impact	PRISMA review
[29]	2025	Hybrid rural system (PV–Wind–Diesel–Battery)	Wind, irradiance, load	COE, NPC	Low CO2 emissions	HOMER + metaheuristics
[30]	2025	Hybrid PV–Diesel (industrial railway)	Load, system sizing	Energy savings	Fossil reduction	Design methodology
[31]	2025	Hybrid PV–Battery–Diesel (rural optimization)	Dispatch, SOC, generation	NPC, COE	Emission comparison	HOMER
[32]	2026	Hybrid PV–Diesel–Battery (off-grid Ethiopia)	Load, solar radiation, sizing	NPC, COE, fuel consumption	Emission reduction	HOMER + PVsyst

presents a comparative analysis of recent studies (2023–2026) focusing on hybrid energy systems that incorporate diesel generation as a backup source. These systems, commonly configured as combinations of photovoltaic (PV) and wind generation, storage, and grid connection, have been extensively researched due to their ability to ensure the reliability of the electricity supply in contexts with high variability in renewable resources or limited electrical infrastructure.

From a technical standpoint, the analyzed studies agree on the consideration of key variables such as solar irradiance, wind speed, load profiles, state of charge (SOC) of batteries, and fuel consumption. These parameters allow for modeling the dynamic behavior of the system and evaluating the interaction between renewable sources and conventional generators. In particular, it is observed that the integration of diesel generators remains essential for ensuring operational stability, especially in isolated scenarios or those with significant intermittency of renewable resources. In economic terms, the most commonly used indicators are the levelized cost of energy (LCOE) and net present cost (NPC) as well as variables related to initial investment (CAPEX), operating costs (OPEX), and fuel consumption. The results show that the incorporation of renewable energy significantly reduces the operating costs associated with diesel use, although economic viability depends largely on factors such as fuel price, the discount rate, and renewable penetration. Recent studies (2024–2026) have also incorporated more advanced approaches, such as energy dispatch optimization and the use of linear programming and metaheuristic algorithms, with the aim of minimizing total system costs.

From an environmental perspective, all studies agree that reducing CO₂ emissions is one of the main benefits of hybrid systems. In particular, configurations that combine PV and diesel achieve significant reductions in emissions compared to systems based exclusively on fossil fuels, with mitigation levels reaching up to 60–90% depending on the level of renewable energy penetration. Likewise, some studies incorporate additional indicators such as the renewable share and health impacts, which reflects a trend toward more comprehensive assessments.

In terms of methodological approaches, a clear evolution in the tools used can be observed. While simulation-based studies and sensitivity analyses using platforms such as HOMER Pro predominated in 2023, subsequent years saw a transition toward more advanced methodologies, including modeling in MATLAB R2014 v8.3/Simulink, energy management strategies (EMSs), artificial intelligence (ANFIS), and multi-criteria optimization. Additionally, the most recent studies incorporate systematic reviews using PRISMA methodologies, which strengthens the scientific rigor of the evidence synthesis.

Finally, the analysis reveals a consistent trend: diesel generators remain an essential component of the energy backup system, but their operation is being progressively optimized to reduce their use and minimize costs and emissions. However, significant research gaps have been identified, particularly regarding the integration of grid-connected hybrid systems with dynamic pricing schemes (such as time-of-use rates), as well as in the development of models that account for specific conditions in local energy markets. Overall, the results show that hybrid PV–diesel systems continue to evolve toward more efficient, intelligent, and sustainable configurations, establishing themselves as a key solution for the energy transition in both off-grid and grid-connected settings.

The consolidated comparative table—hybrid energy systems (diesel-based, 2023–2026)—shows that most recent studies focus on optimizing energy costs (LCOE, NPC) and reducing emissions through hybrid configurations that integrate PV–diesel–battery combination or those with storage, prioritizing the increase in the renewable share and system reliability. However, these approaches generally overlook the influence of actual time-of-use rate structures and, in particular, the issue of peak demand charges, which represent a critical component in electrical systems under rate schemes such as the Hourly Medium-Voltage Peak Demand (GDMT) system. In this context, this study stands out by proposing a strategy focused on mitigating demand peaks through the operational dispatch of a diesel generator without a storage system, specifically aimed at reducing demand charges during peak periods, rather than merely maximizing renewable energy penetration. This approach introduces a significant conceptual shift: the diesel generator ceases to be merely an energy backup and becomes a tool for active tariff management. The scientific contribution lies in the integration of a technical–economic optimization model coupled with hourly electricity tariffs, evaluated in a real-world case study of a university campus on the Iberian Peninsula, where actual load profiles, local climatic conditions, and specific tariff structures are taken into account. This enables the development of a replicable methodology that links the operation of hybrid systems with demand-side cost-reduction strategies, providing quantitative evidence on the viability of PV–grid–diesel schemes without batteries in institutional settings—an approach that has received little attention in the recent literature.

Figure 1 shows the percentage distribution of global CO₂ emissions among the 15 countries with the largest contribution worldwide. In general, China, the United States, and the European bloc account for the largest share of emissions, reflecting their historical and current role in fossil fuel-intensive energy systems. In particular, the countries highlighted in yellow—the United States, Canada, Mexico, and Brazil—represent a heterogeneous group in both economic and demographic terms, but they share the characteristic of being part of the group of intermediate emitters in global weight. During 2023, Mexico experienced a significant increase in its carbon dioxide (CO₂) emissions, reaching 487,094 megatons, which represents an increase of 4.54% over the previous year. This behavior places the country among the leading global emitters among the 184 nations analyzed [33]. The per capita analysis also showed an increase, registering 3.52 tons of CO₂ per inhabitant in 2023, while the carbon intensity of the economic product remained stable at 0.17 kg per $1000 of GDP, thus maintaining environmental efficiency without notable variations [34]. Although historical data show a sustained decline in total emissions since 2013, data from the last five years reveal an upward trend, signaling a setback in sustainability and climate mitigation for the country [35].

This study presents a technical, economic, and environmental analysis of a hybrid renewable energy system (SHER), based on the modeling of four operational scenarios representing different levels of coverage (30%, 50%, 70%, and 100%) of the energy demand associated with a commercial facility located on a university campus in the Yucatán Peninsula. The system analyzed consists of a photovoltaic array interconnected to the electrical grid and a diesel generator, whose joint operation allows for the examination of supply strategies under real tariff conditions. The evaluation was carried out considering a medium-voltage tariff scheme with hourly control of energy dispatch, characterized by the application of differentiated charges according to base, intermediate, and peak periods, as well as seasonal variations between summer and winter periods. This tariff system simultaneously incorporates hourly energy consumption and contracted demand at medium voltage, which allows for an accurate reproduction of the economic behavior of the supply and its interaction with the operation of the hybrid system [36].

The analysis and optimization of hybrid renewable energy systems (HERSs) require advanced tools that allow the technical, economic, and environmental performance of different configurations to be evaluated under real operating conditions. In this context, the HOMER Pro software has established itself as one of the most robust platforms for simulating and sizing hybrid photovoltaic–diesel–grid systems, due to its ability to integrate hourly load profiles, variations in renewable resource availability, and complex tariff schemes. This research relies on this tool to compare different levels of renewable generation participation in a university facility, evaluating scenarios with different degrees of photovoltaic penetration and analyzing their economic and environmental implications. It also considers aspects specific to medium-voltage electrical systems, which incorporate variable charges related to energy consumption, peak demand, and time periods, as described in the current tariff structure for this type of facility. This approach allows for a realistic representation of the behavior of the hybrid system and an assessment of its viability in institutional environments with high energy requirements and variable load patterns [37].

1.1. General Approach to the Systematic Review

This study adopted a structured systematic review strategy under the PICOT framework, with the aim of identifying, analyzing, and synthesizing scientific literature related to hybrid renewable energy systems, particularly in university contexts, using simulation tools such as HOMER Pro, as well as energy sizing and optimization approaches. The use of the PICOT framework allowed us to precisely define the key elements of the search, ensuring traceability, transparency, and replicability at all stages of the process [38,39]. The review was conducted following the methodological guidelines recommended by the scientific literature for systematic studies, applying progressive filters for identification, screening, eligibility, and inclusion, as detailed in the previously developed PRISMA textual diagrams. This approach allowed us not only to evaluate the thematic relevance of the articles, but also to identify temporal trends based on the years of publication and the use of keywords related to university, HOMER Pro, and sizing (sizing, optimal sizing, dimensioning, capacity design, capacity planning, system sizing).

1.2. Search Strategy and Databases

Search strings were constructed using synonyms and equivalent terms, with the Boolean operators AND and OR, following standards for systematic searches in engineering and energy sciences. Keywords were selected based on previous studies analyzing the optimization and design of hybrid systems. Queries were performed in multiple high-impact scientific databases, including IEEE Xplore with 254 documents, prioritizing peer-reviewed articles published between 2020 and 2026, a period in which there has been an intensification of research on renewable integration, simulation, and energy modeling.

1.3. PICOT Process Applied to the Review

The PICOT framework was used to define the search criteria and build a robust conceptual filter to guide the identification of relevant literature.

Table 2. Description using PICOT search frameworks.

Acronyms	Description
P—Population	Hybrid renewable energy systems (PV-diesel, grid, microgrids, energy campus system).
I—Intervention	Application of simulation tools: primarily Homer Pro and associated platforms.
C—Comparator	Traditional system based on fossil fuels or non-optimized configuration.
O—Out (Results)	Sizing optimization, technical–economic analysis, emission reduction, energy planning strategies.
T—Temporal Horizon	Articles published between 2020 and 2026, a period in which an intensification in the research on renewable energy, microgrids, and advanced simulation methodologies is observed.

presents the PICOT structure used.

1.4. Identification and Filtering Through Keywords

The articles were filtered according to three main thematic categories:

University context: Studies mentioning university, campus, or educational institutions in their screening phase were verified to ensure that each study included energy analysis, modeling, or assessment applicable to academic facilities. All articles met this criterion and therefore proceeded to the eligibility phase. In the eligibility phase, duplicate studies or those with insufficient information were eliminated, resulting in a total of 13 valid studies, distributed evenly between 2020 and 2023 (three per year), with a notable decrease in 2024, in which only one article was recorded. After applying the thematic filter, 13 articles published between 2020 and 2024 were identified.

Use of HOMER Pro: articles with explicit reference to HOMER Pro or During the filtering phase, materials that mentioned HOMER Pro only tangentially were eliminated. All remaining articles showed clear methodological use and therefore proceeded to eligibility. Eligibility criteria were applied regarding:

(1)

Clarity in the simulation process,

(2)

Description of technical parameters,

(3)

Direct relationship with hybrid systems.

After this evaluation, the 17 articles were considered valid for analysis. The temporal distribution shows a notable evolution:

• A low level in 2021,
• A substantial increase in 2023 (five articles),
• High and consistent values in 2024–2025 (three and four articles, respectively)
• And even in 2026, there is evidence of continued growth in the use of HOMER Pro for energy studies.

This pattern shows that HOMER Pro has established itself as the dominant tool in energy modeling and optimization work.

Sizing: Research that includes terms such as sizing, optimal sizing, dimensioning, capacity design, capacity planning, and system sizing. The identification process detected 29 articles related to the sizing of hybrid energy systems. This category showed the fastest growth of the three lines analyzed. The temporal distribution shows significant progress:

• Sustained growth between 2020 and 2022;
• A turning point in 2024 with seven articles;
• A peak in 2025 with nine articles;
• Even in 2026, there are already two early publications.

Figure 2 shows a circular diagram of three concentric rings that summarizes the temporal distribution of the articles included in the systematic review, grouped according to three thematic lines identified from the PRISM analysis: (1) applied research in university contexts (upper ring), (2) studies using the HOMER Pro tool (middle ring), and (3) articles focused on the sizing and optimization of hybrid systems (lower ring). The results show that the field of study has evolved from approaches focused on specific contexts, such as university environments, to more robust methodological approaches based on modeling, optimization, and the integration of specialized computational tools. This transition reflects not only a maturation of the lines of research, but also the need for more accurate and replicable methodologies for the evaluation of hybrid systems in different application scenarios.

The methodology described is directly related to the general objective set out in the introduction to this study: to understand the evolution of research on hybrid systems applied to university environments and energy optimization methodologies. The use of PICOT, together with thematic classification, allows us to observe how studies on HOMER Pro, energy modeling, and optimal sizing have increased in publication frequency in recent years, with 2023–2025 standing out as the period of greatest scientific productivity. This methodological approach provides a clear conceptual map of the state of the art, underpins the selection of studies included, and establishes solid criteria for subsequent quantitative and qualitative analysis.

2. Methodology

2.1. General Method of the Hybrid System on Campus

The methodology used in this study was structured with the purpose of developing a techno-economic–environmental analysis for the optimal sizing of a hybrid renewable energy system (HRES), using HOMER Pro software as the main simulation tool. Figure 3 View.

The process began with the definition of the case study, corresponding to the electricity supply of a university campus located on the Yucatan Peninsula in Mexico. This stage included the preliminary collection of relevant contextual information to adequately characterize the existing electrical system. Subsequently, an electrical audit was carried out to determine current operational behavior and collect the parameters necessary to construct the system’s load profile. This characterization made it possible to establish the temporal variations in electricity consumption and formed the basis for the subsequent data integration and modeling processes.

With the load profile defined, the technical, geographical, and economic data were integrated. This phase incorporated the renewable energy resources available at the study location, the climatic conditions, investment and operating costs, and the parameters of the current tariff system. This tariff scheme corresponds to a medium-voltage service with a time-of-use structure, which applies differentiated charges according to base, intermediate, and peak periods, in addition to considering seasonal variations and the user’s contracted demand.

At the same time, the technical properties of each HRES component were specified, including the system topology, diesel generator characteristics, grid electrical parameters, photovoltaic array properties, and converter and controller capacities and efficiencies. These elements were integrated into the HOMER Pro V3.14 and PVsys V7.4 simulation environment, where the base configurations and variation ranges for sensitivity analyses were defined.

Once the parameterization was complete, the HRES was dimensioned, considering four levels of renewable penetration: 30%, 50%, 70%, and 100% of the campus’s electricity demand. Simultaneously, the generator was specifically dimensioned, evaluating its operational contribution in each scenario and its interaction with the other components of the hybrid system. From the models generated, the technical–economic–environmental results were obtained, including the net present cost, levelized cost of energy, CO₂ emissions, fuel consumption, and optimization indicators associated with the joint operation of the system. A sensitivity analysis was also developed to evaluate the impact of the most critical economic variables on the HRES. Finally, the results were analyzed comparatively to determine the most efficient configuration in terms of cost reduction and performance, environmental impact, and operational reliability of the electricity supply.

2.2. Case Study

The case study is located in the central-eastern region of the state of Yucatán, selected for three fundamental factors: the high availability of solar resources characteristic of the area, the presence of continuous electricity demand during daylight hours associated with administrative and infrastructure activities, and operation under a medium-voltage tariff system with a time structure, which allows for accurate assessment of the economic impact of the proposed hybrid system. The institution has approximately 150 employees and 1285 students enrolled in 2023, operating from 7:00 a.m. to 7:00 p.m., Monday through Saturday, in a physical complex consisting of eight buildings, two sports fields, a parking lot, and an environmental unit. In addition, the area has physical spaces suitable for the installation of solar systems and future expansions. According to National Institute of Statistics and Geography (INEGI) data, the area is experiencing significant urban growth, with a population of over 74,000 inhabitants and a territorial expansion of 2.70% compared to the state, conditions that justify the need to explore sustainable and highly efficient energy solutions [40,41].

The geographical location of this region offers particularly favorable conditions for photovoltaic generation. Information from satellite sources, such as the NASA/POWER system, confirms high and stable levels of solar irradiance throughout the year, as well as average temperatures within ranges suitable for ensuring efficient performance of photovoltaic modules. The data on global horizontal irradiance (GHI), monthly and annual solar radiation, and the average, maximum, and minimum ambient temperatures are collected in

Table 3. Meteorological data for the study area in Yucatán, Mexico.

Month	Irradiation Latitude (kWh/m2/day)	Horizontal Irradiation (kWh/m2/day)	Horizontal Irradiance (W/m2)	Latitude Irradiance (W/m2)	Average Temp (°C)	Max Temp (°C)	Min Temp (°C)
Jan	5.0	4.2	177.2	210.9	24.4	34.1	13.2
Feb	5.6	5.0	209.8	235.6	24.3	35.7	13.5
Mar	6.1	5.8	243.3	255.3	27.3	39.6	14.6
Apr	6.3	6.4	270.5	265.1	28.8	39.6	15.8
May	6.0	6.5	271.0	252.3	30.2	40.1	19.8
Jun	4.9	5.3	223.9	204.8	28.6	40.6	21.7
Jul	4.9	5.3	223.5	206.8	27.3	34.5	22.4
Aug	5.1	5.4	225.2	216.4	27.0	31.4	21.8
Sep	5.3	5.2	219.3	223.4	26.9	31.7	20.5
Oct	5.3	4.8	202.3	221.2	25.8	32.1	16.7
Nov	5.0	4.3	180.9	212.4	25.3	33.5	15.7
Dec	4.6	3.8	160.9	193.6	23.8	34.5	14.0

. Analysis of these parameters allows us to determine the energy potential of the area for photovoltaic systems and evaluate the stability of the renewable resource.

2.3. Historical Analysis of Energy Demand and Tariff Structure

Official electricity billing records for the period January–December 2023 were analyzed as part of the demand characterization process. Each bill included information broken down by tariff period, differentiating between energy consumption (kWh) and maximum demand (kW) in the three time windows defined by the electricity system’s operating scheme: base, intermediate, and peak periods. Likewise, the economic charges associated with supply, distribution, transmission, generation, and capacity services were evaluated, along with those established by the entities responsible for regulating and administering the electricity market, which include compensation and operation mechanisms for the national energy system. Figure 4 presents the percentage breakdown of the components that make up the annual electricity bill for a medium-voltage service with a time-of-use structure. This analysis allows us to identify how the total cost of supply is distributed throughout the year, expressed in percentage terms for each of the tariff items defined by the entities responsible for the administration of the electricity system. It can be seen that 60.5% of the total cost is directly associated with the energy consumed, which shows that the main economic impact comes from the sustained electricity demand during the intermediate period, which alone accounts for 54.4% of the annual consumption. Capacity charges (16.48%) and distribution charges (14.89%) are the second most important components within the tariff structure, reflecting costs related to infrastructure, service availability, and operational support for the system. In contrast, items related to transmission (7.12%), supply (0.48%), and fees established by the agencies responsible for regulating and coordinating the electricity market (0.27% and 0.25%) have a relatively smaller share, although they are essential to ensure the continuous operation and balance of the energy system. Furthermore, internal analysis of the tariff periods confirms a predominantly daytime consumption pattern: the intermediate period accounts for the largest share of energy use at 54.4%, while the base (3.3%) and peak (2.8%) periods show significantly lower contributions. This behavior is characteristic of facilities whose operation depends mainly on academic, administrative, or institutional activities carried out during the day.

Figure 5 shows monthly consumption in kWh, which varies significantly throughout the year, with consumption clearly peaking during the intermediate period, corresponding to daytime hours. Of all the months, May is identified as the most critical, reaching a total consumption of 58,879 kWh. In Figure 6, the highest maximum demand value was 299 kW, recorded during the month of May, coinciding with high-thermal-load weather conditions and intensive use of air conditioning systems. This value directly influences the charges associated with the infrastructure and operation of the electrical system, which is reflected in the total monthly billing cost.

The graph in Figure 7 shows the percentage distribution of electricity consumption and the associated billing cost for the three tariff periods (base, iintermediate, and peak) throughout each month of the year. The blue, orange, and yellow blocks represent the percentage of consumption during base, intermediate, and peak hours, respectively, while the gray, green, and light blue segments show the percentage corresponding to the cost billed for those same hours. It can be seen that, in all months, the Intermediate period accounts for most of the energy consumption, consistently exceeding 80%, which reflects a predominantly daytime usage pattern.

Figure 8 analyzes the monthly costs associated with generation times, revealing different patterns throughout the year. In the case of base generation, the highest cost was recorded during the month of March, with a total of $5112.62 million MNM. For intermediate generation, which is the component with the greatest economic weight in the billing structure, the month of May presented the highest value, reaching $98,330.30 million MNM. Peak generation showed its highest cost in November, with $3269.58 million MNM, standing out as the period in which this component had its highest relative contribution. When considering the total monthly cost, which includes the three generation periods, May again shows the highest value, with an amount of $105,425.49, confirming that this month has the highest economic burden of the year. This behavior is associated with both consumption levels and seasonal variations typical of the region, which increase energy demand and, therefore, the final cost of supply.

Figure 9 shows the monthly costs associated with the transmission component, with the highest value recorded in May, with a total of $10,983.25 MNM, making it the period with the greatest economic impact in this regard during the year evaluated.

Figure 10 shows the percentage of cost data by distribution and capacity. The highest distribution value was recorded in May, with a cost of $13,373.01, followed by September and October. Capacity also had its highest cost in May, reaching a total of $37,289.06, which represents the highest point observed throughout the year. This increase is consistent with the increase in peak demand recorded in that period, given that capacity charges are directly related to the power requirements that the electrical system must guarantee.

The financial analysis indicates that the annual cost of electricity supply amounts to MXN 1,313,579.70, distributed among the various components defined by the medium-voltage tariff structure with time control. Among these charges, the largest economic contributions come from items related to generation, capacity, and transmission, which highlights the importance of reducing peak demand and promoting energy displacement through on-site self-generation (see

Table 4. Totals costs.

Month	Total Fixed Cost	Total Hourly Energy Cost	Total Variable Energy Cost	Total Demand Cost	$\mathbf{\sum }\mathbf{T}\mathbf{o}\mathbf{t}\mathbf{a}\mathbf{l}$ Final Bill (Without VAT)
Jan	$441.8	$47,172.8	$5099.2	$27,115.3	$79,829.3
Feb	$441.8	$59,235.0	$6377.7	$29,792.7	$95,847.5
Mar	$441.8	$89,514.7	$9406.4	$29,583.5	$128,946.5
Apr	$441.8	$82,487.2	$8685.5	$20,382.7	$111,997.3
May	$441.8	$105,425.4	$10,983.2	$51,117.4	$167,968.0
Jun	$441.8	$73,618.7	$7788.6	$18,509.2	$100,358.5
Jul	$441.8	$41,359.8	$4597.3	$18,466.4	$64,865.4
Aug	$441.8	$74,192.0	$7835.4	$17,451.9	$99,921.3
Sep	$441.8	$98,316.4	$10,253.2	$21,778.3	$130,789.8
Oct	$441.8	$104,201.3	$10,853.7	$22,109.1	$137,606.0
Nov	$441.8	$76,145.1	$8032.9	$28,394.9	$113,014.8
Dec	$441.8	$47,702.3	$5202.0	$29,088.4	$82,434.7
Totals	$5302.4	$899,371.2	$95,115.6	$313,790.3	$1,313,579.70

).

These results allow us to conclude that the energy profile of the facility presents favorable conditions for the incorporation of a hybrid system based on photovoltaic generation with diesel backup and connection to the grid. The high correlation between the available solar resource and the predominant consumption during the intermediate period suggests a high potential for self-consumption and, consequently, a significant reduction in both billed energy and charges associated with peak demand.

Based on the load profile characterization and tariff analysis for the base, intermediate, and peak periods, different energy configurations were evaluated, integrating a photovoltaic system, a diesel generator, and the electrical grid. The analysis showed that the campus’s highest demand is concentrated in the intermediate period, coinciding with the availability of solar resources, while the peak period, although short in duration, has a decisive impact on demand charges. Given this behavior, the selected hybrid topology, consisting of a grid-connected photovoltaic system to cover mainly intermediate demand, a Hyundai HY36kW diesel generator (Made-in-China, Hyundai Power Equipment global. Distributor, The Garden Shop Maya (Mexico City, Mexico)) to mitigate peak demand, and the grid as the base supply, is the most suitable alternative from a technical and economic point of view; see the graph in Figure 11. The system simulation in HOMER Pro used a representative profile of a commercial–industrial installation on the Yucatan Peninsula in Mexico, observing typical daytime demand behavior, with reduced nighttime consumption (10–20 kW) and peaks between 70 and 100 kW around midday, coinciding with the highest availability of solar irradiance. This pattern favors a high photovoltaic contribution and reduces both the energy purchased from the grid and the need for prolonged operation of the diesel generator. On a monthly basis, higher demand was identified between March and September, associated with operational increases and the thermal load derived from hot weather conditions, reaching values of up to 250 kW, while in winter the load decreases to ranges of 120–150 kW. In this context, the diesel generator strategically operates as a peak suppression system during peak hours, replacing high-cost grid energy and avoiding peak demand records that increase capacity and distribution charges. This strategy significantly reduces the annual costs associated with electricity consumption. It simultaneously reduces the expenses associated with peak kWh and maximum kW recorded, which is reflected in a direct decrease in the total annual cost of energy.

2.4. Sizing for the Diesel Generator

A detailed characterization of the annual, monthly, and hourly load is carried out in order to identify seasonal patterns and demand peaks that directly influence the sizing of the system. In this context, the sizing and proper operation of the generator are essential to ensure energy efficiency and extend its useful life. According to ISO 8528 and manufacturer specifications, generator sets are classified into Standby, Prime, and Continuous modes, each with specific operating limits [42]. In backup applications, the recommended operation does not exceed 200 h/year, with a capacity factor (CF ≈ 2.3%), and may exceptionally reach 500 h/year (CF ≈ 5.7%) [43]. For frequent operation, it is recommended not to exceed 70% of the average rated power, maintaining 100% only for short periods, while the annual CF can vary between 25% and 70% depending on interaction with renewable sources [44]. Additionally, manufacturers warn that sustained operation below 30–40% causes soot accumulation and incomplete combustion, increasing maintenance requirements [45]. These restrictions determine the operational role of the generator in hybrid systems, where its exclusive use as backup implies CFs of less than 5%, while in continuous support schemes it can reach up to 70%, depending on the design and renewable penetration of the system. The analysis of peak consumption showed monthly variations of average power between 8.91 kW and 23.26 kW, see

Table 5. Parameters for peak electricity consumption and fuel.

Month	Electricity Consumption (kWh)	Hours of Consumption	Average kW (Hour)	Load %	Interpolated Consumption (L/h)	Estimated Monthly Consumption (L/Month)
Jan	1869	100	18.69	52%	5.1	515.3
Feb	1705	92	18.53	51%	5.1	470.8
Mar	1872	94	19.91	55%	5.4	510
Apr	855	96	8.91	25%	2.9	286
May	1070	46	23.26	65%	6.1	283.7
Jun	764	40	19.10	53%	5.2	209.7
Jul	722	46	15.70	44%	4.4	206.4
Aug	788	44	17.91	50%	4.9	219.1
Sep	768	42	18.29	51%	5.0	212.6
Oct	856	46	18.61	52%	5.1	236.2
Nov	1570	94	16.70	46%	4.7	442.8
Dec	1466	96	15.27	42%	4.3	421.7
Total	14,305	836	-	-	-	4014.8

, evidencing a marked seasonality that influences the selection of backup equipment. The month with the highest demand was May, with 23.26 kW (65% load), while April recorded the minimum value (25%).

Based on the criteria of efficiency, reliability, and service life, an optimal load factor of 75% was adopted [46,47] which estimated a required power of 31 kW for the generator. However, to ensure reserve capacity and safe operation, the commercial 36 kW (45 kVA) model HYUNDAI HYEG36KW was selected, which is suitable for meeting the identified demands. The manufacturer’s technical data sheet provided the generator’s specific consumption curve, summarized in

Table 6. Fuel consumption data of the generator to be used: HYEGE36KW HYUNDAI.

Capacity	kVA	kW	Li/h
100%	45	36	9
75%	33.75	27	7
50%	22.5	18	5
25%	11.25	9	3

, which shows that diesel consumption increases proportionally to the load level, from 3 L/h at 25% to 9 L/h at 100%. By correlating this curve with the monthly demand profile and the actual hours of operation during peak hours, it was possible to model the actual fuel consumption for each month.

2.5. Correlation Between Fuel Consumption and Load Percentage

The trend shows an estimated annual consumption of 4015 L of diesel in the baseline scenario, corresponding to the energy supplied by the generator during periods of the greatest tariff impact.

$${COV}_{xy}={\displaystyle \frac{\sum (x_i-\overline{x})(y_i-\overline{y})}{n-1}}$$

(1)

$$s_x^2={\displaystyle \frac{\sum {(x_i-\overline{x})}^2}{n-1}}$$

(2)

$$\widehat{y}={\beta }_0+{\beta }_{1}x$$

(3)

$${\beta }_1={\displaystyle \frac{{COV}_{xy}}{s_x^2}}$$

(4)

$${\beta }_0=\overline{y}-{\beta }_{1}x$$

(5)

From Equations (4) and (5) we obtain the model for Equation (3).

$$\widehat{y}=1+8x$$

(6)

Equation (6) is used to generate the data to obtain the load percentages for each month shown in Table 5.

Characterizing the generator using indicators such as electrical efficiency and specific fuel consumption (SFC) is essential for evaluating its contribution within the hybrid system. Efficiency allows us to determine what proportion of diesel energy is converted into useful electricity, while SFC directly relates fuel consumption to the energy produced. These parameters allow for adequate modeling of its operation, estimation of the annual consumption and costs, and analysis of its interaction with the PV–generator–grid system. The results obtained allow for the identification of the optimal operating conditions and the definition of strategies that prioritize the use of photovoltaic energy, contributing to a reduction in fossil fuels, economic optimization, and the reduction of the environmental impact associated with thermal generation.

Assumptions. The lower heating value (LHV) of diesel was set at 35.8 MJ/L (typical value).

$$\mathrm{Conversion}: 1\mathrm{M}\mathrm{J}/\mathrm{h}={\displaystyle \frac{1}{3.6}}\mathrm{k}\mathrm{W}$$

Thermal energy provided by the fuel (kW term):

$$P_{comb}\left(\mathrm{k}\mathrm{W}\right)=\dot{m_f}(\mathrm{L}/\mathrm{h})\times PCI(\mathrm{M}\mathrm{J}/\mathrm{L})\times {\displaystyle \frac{1}{3.6}}$$

(7)

Electrical efficiency (%):

$${\eta }_{el}={\displaystyle \frac{P_e}{P_{comb}}}\times 100$$

(8)

Specific fuel consumption (SFC) in L/kWh:

$$\mathrm{S}\mathrm{F}\mathrm{C}={\displaystyle \frac{\dot{m_f}(\mathrm{L}/\mathrm{h})}{P_e\left(\mathrm{k}\mathrm{W}\right)}}$$

(9)

Table 7. Characterization of the behavior of the generator by means of performance parameters.

Percentage (%)	Electric Power (kW)	Consumption (L/h)	Fuel Energy (kW Therm.)	Electrical Efficiency (%)	SFC (L/kWh)
100	36	9	89.5	40.2	0.25
75	27	7	69.6	38.7	0.25
50	18	5	49.7	36.2	0.27
25	9	3	29.8	30.1	0.33

summarizes the technical and economic parameters used to model the operation of the HYUNDAI HYEG36KW generator within the PV–diesel–grid hybrid system.

Figure 12 shows the relationship between average monthly peak power and estimated generator fuel consumption, showing a decrease in consumption during the summer even with high demand. Evaluation based on electrical efficiency and specific fuel consumption (SFC) allows for the quantification of the useful energy generated and the diesel required, which is essential for modeling the operation, estimating the consumption and costs, and identifying strategies that favor greater photovoltaic participation. In HOMER Pro, the generator was configured with investment, replacement, and operating costs, a useful life of 25,000 h, a minimum load of 30%, a fuel price of $22/L, and its actual consumption curve, allowing its performance within the hybrid system to be accurately represented.

2.6. Generator Configuration in Homer Pro

The diesel generator configuration in HOMER Pro was developed following technical criteria established in international standards and manufacturer specifications in order to accurately represent its operation within the hybrid system [46,47,48]. The essential parameters of the equipment were incorporated, including its rated power, capital and replacement costs, operating costs, and useful life expressed in hours, which allowed for an accurate estimate of the component’s lifecycle. Likewise, the fuel consumption characteristic curve was integrated to model the load–consumption relationship with high fidelity [49]. A minimum load of 30% was established to avoid inefficiencies associated with incomplete combustion, and the cost of diesel was set according to its regional value. In addition, a dispatch strategy was defined using the Generator Schedule module, delimiting periods of operation and shutdown according to demand conditions and economic criteria. This procedure ensures that the behavior of the generator within the model accurately represents its operation under real conditions. The incorporation of dispatch strategies, combustion curves, and cost parameters into the model not only allows the physical operation of the generator to be reproduced, but is also an essential methodological requirement for ensuring the validity of the analysis in hybrid-system studies. From an academic perspective, establishing a controlled dispatch scheme allows for analysis of the generator’s participation at specific times of the day or under critical tariff conditions, especially in systems where renewable energy and the electrical grid coexist with variable loads. Likewise, the combustion curve is essential for estimating actual diesel consumption, since thermal generators do not operate linearly and their efficiency changes markedly between low loads and loads close to the nominal value, which directly affects costs and emissions. Figure 13 shows the main configurations used in HOMER Pro to represent this behavior.

2.7. Sizing for the Photovoltaic System

The sizing of the grid-connected photovoltaic system must cover the maximum daily demand recorded on the university campus during the month of May, being conservative in the sizing, which reached a value of 1984.6 kWh/day. This calculation was based on the minimum monthly solar irradiance at the site (3.8635 kWh/m²·day) and an overall system loss factor (PR = 0.70), parameters that allow for a conservative estimate of the power required under unfavorable operating conditions. Based on these values, it was determined that the required photovoltaic power is in the order of 746 kWp.

The sizing process began with the collection and detailed analysis of energy consumption for the intermediate period of the medium-voltage tariff system with a time structure, using the billing records for the twelve months evaluated. This analysis made it possible to identify the usage patterns, distribution of consumption by time period, and maximum demand values that characterize the annual behavior of the installation.

Table 8. Electricity consumption parameters at intermediate rates.

Month	Monthly Consumption (kWh)	Total Hours Intermediate Rate	Average Load per Hour (kW)	Maximum Hours of Intermediate Rate per Day	Average Consumption kWh/d Max
Jan	21,169	402	52.6	14	737.2
Feb	27,625	374	73.8	14	1034.0
Mar	43,583	388	112.3	14	1572.5
Apr	41,285	388	106.4	14	1489.6
May	53,234	456	116.7	17	1984.6
Jun	36,739	430	85.4	17	1452.4
Jul	19,299	456	42.3	17	719.4
Aug	37,092	457	81.1	17	1379.7
Sep	49,970	429	116.4	17	1980.1
Oct	52,911	456	116.0	17	1972.5
Nov	37,103	388	95.6	14	1338.7
Dec	21,940	394	55.6	14	779.5

shows the monthly consumption, number of hours associated with each period, and resulting average load for each month. To calculate the sizing, the maximum hours recorded monthly were considered, assuming that they represent the most critical operating scenario that must be addressed [50,51]. Finally, the average consumption corresponding to these critical hours was estimated, in order to establish a conservative and representative basis for the system design.

The solar resource is characterized by obtaining data on global irradiation, temperature, and peak solar hours (Table 2) in order to determine the site’s energy availability and identify the months with the highest and lowest potential generation. The table shows three criteria for estimating kWp using the peak solar hours (HSP) for each month, under a conservative criterion (most critical case) considering the minimum monthly solar irradiation of the site (3.8635 kWh/m²·day) and combining this with a loss factor of 0.7. The approximate number of panels for these three cases is obtained, taking the commercial value of the 555 W panel and the value of 1984.60 kWh/d for the month of May. Using Equations (10) and (11). The calculation is performed each month, using the average HPS value for the month with the minimum HSP value (3.8) and the losses factored (0.7).

Table 9. Summary of consumption data by intermediate hours, peak solar hours, and estimated number of panels.

Month	kWh/d Max.	HSP	kWp HSP	kWp HSP Minimum	kW HSP 70%	Number of Panels HSP	Number of Panels HSP Min.	Number of Panels HSP Losses
Jan	737.2	4.3	173.4	194.0	277.2	315.2	352.7	503.9
Feb	1034.1	5.0	205.4	272.1	388.8	373.4	494.8	706.8
Mar	1572.6	5.8	269.2	413.8	591.2	489.5	752.4	1074.9
Apr	1489.7	6.5	229.4	392.0	560.0	417.1	712.8	1018.2
May	1984.6	6.5	305.1	522.3	746.1	554.6	949.6	1356.5
un	1452.5	5.4	270.3	382.2	546.0	491.4	695.0	992.8
Jul	719.5	5.4	134.1	189.3	270.5	243.8	344.3	491.8
Aug	1379.8	5.4	255.3	363.1	518.7	464.1	660.2	943.1
Sep	1980.2	5.3	376.2	521.1	744.4	684.0	947.4	1353.5
Oct	1972.6	4.9	406.1	519.1	741.6	738.4	943.8	1348.3
Nov	1338.8	4.3	308.3	352.3	503.3	560.5	640.6	915.1
Dec	779.6	3.9	201.8	205.2	293.1	366.9	373.0	532.9

summarizes the calculated data.

$$Nominal power of the Panels={\displaystyle \frac{\frac{Daily Consumption \mathrm{W}\mathrm{h}/\mathrm{d}\mathrm{a}\mathrm{y}\mathrm{s}}{Hours of sunlight}}{0.7}}$$

(10)

$$Number of panels={\displaystyle \frac{Nominal power of the panel}{Panel power}}$$

(11)

2.8. Corrected Temperature Calculations

The parameters for estimating the corrected temperatures are described in Table 10.

$$T_{cmax}=T_{a}max+\left({\displaystyle \frac{NOCT-20}{800}}\right)\ast G_{max}$$

(12)

$$T_{cmin}=T_{a}min+\left({\displaystyle \frac{NOCT-20}{800}}\right)\ast G_{min}$$

(13)

• $T_{cmax}=\mathrm{C}\mathrm{o}\mathrm{r}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{e}\mathrm{d} \mathrm{m}\mathrm{a}\mathrm{x}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m} \mathrm{t}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e} (°\mathrm{C})$;
• $T_{cmin}=\mathrm{C}\mathrm{o}\mathrm{r}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{t}\mathrm{e}\mathrm{d} \mathrm{m}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m} \mathrm{t}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e} (°\mathrm{C})$;
• $T_{a}max=\mathrm{M}\mathrm{a}\mathrm{x}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m} \mathrm{a}\mathrm{m}\mathrm{b}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{t} \mathrm{t}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e} (°\mathrm{C})$;
• $T_{a}min=\mathrm{N}\mathrm{o}\mathrm{m}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{l} \mathrm{c}\mathrm{e}\mathrm{l}\mathrm{l} \mathrm{o}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{t}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e} (°\mathrm{C})$;
• $NOCT=\mathrm{N}\mathrm{o}\mathrm{m}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{l} \mathrm{o}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{t}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e} \mathrm{o}\mathrm{f} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{c}\mathrm{e}\mathrm{l}\mathrm{l} (°\mathrm{C})$.

Table 10. Parameters for temperature correction.

Parameter	Value
Maximum recorded ambient temperature (Tamax)	40.66 °C
Minimum recorded ambient temperature (Tamin)	13.21 °C
NOCT	45 °C
Maximum irradiance (Gmax)	265.11 W/m2
Minimum irradiance (Gmin)	193.6 W/m2

Table 10. Parameters for temperature correction.

Parameter	Value
Maximum recorded ambient temperature (Tamax)	40.66 °C
Minimum recorded ambient temperature (Tamin)	13.21 °C
NOCT	45 °C
Maximum irradiance (Gmax)	265.11 W/m2
Minimum irradiance (Gmin)	193.6 W/m2

Using the corrected temperature data, $T_{cmax}=48.8 °\mathrm{C}$ and $T_{cmin}=19.2 °\mathrm{C}$, the corrected electrical parameters $V_{octc}$, $I_{sctc}$ and $V_{octc}$, $I_{sctc}$ are calculated using the values from the tables in the 555 W panel characteristic sheet. Table 11 shows the STC electrical characteristics, and Table 12 shows the temperature coefficients. By applying the corresponding equations, the corrected voltages, currents, and power for the corrected temperatures are obtained.

$$V_{octc}=\left\{1+\left[{\displaystyle \frac{\left[\left(T_c-25 °\mathrm{C}\right)\ast \alpha \right]}{100}}\right]\right\}\ast V_{OC}$$

(14)

$$I_{sctc}=\left\{1+\left[{\displaystyle \frac{\left[\left(T_c-25 °\mathrm{C}\right)\ast \beta \right]}{100}}\right]\right\}\ast I_{SC}$$

(15)

$$P_T=\left\{1+\left[{\displaystyle \frac{\left[\left(T_c-25 °\mathrm{C}\right)\ast \gamma \right]}{100}}\right]\right\}\ast P_{max}$$

(16)

• $V_{oc}min=46.43 \mathrm{V} \mathrm{w}\mathrm{i}\mathrm{t}\mathrm{h} \mathrm{m}\mathrm{a}\mathrm{x}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m} \mathrm{t}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e}$;
• $V_{oc}max=50.73 \mathrm{V} \mathrm{w}\mathrm{i}\mathrm{t}\mathrm{h} \mathrm{m}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m} \mathrm{t}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e}$;
• $I_{sc}max=14.2 \mathrm{A} \mathrm{a}\mathrm{t} \mathrm{m}\mathrm{a}\mathrm{x}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m} \mathrm{t}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e}$;
• $I_{sc}min=14 \mathrm{A} \mathrm{a}\mathrm{t} \mathrm{m}\mathrm{a}\mathrm{x}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m} \mathrm{t}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e}$;
• $P_{Tc}min=507.15 \mathrm{W} \mathrm{a}\mathrm{t} \mathrm{m}\mathrm{a}\mathrm{x}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m} \mathrm{t}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e}$;
• $P_{Tc}max=566.46 \mathrm{W} \mathrm{a}\mathrm{t} \mathrm{m}\mathrm{a}\mathrm{x}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m} \mathrm{t}\mathrm{e}\mathrm{m}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{u}\mathrm{r}\mathrm{e}$.

Table 11. STC electrical characteristics.

Parameter	Value	Unit	Abbreviation
Output Power	555	W	Pmax
Output Power Tolerance	0/+5	W	Δpmax
Module Efficiency	21.48	%	Hm
Voltage at Pmax	42.24	V	Vmpp
Current at Pmax	13.14	A	Impp
Open-Circuit Voltage	49.9	V	Voc
Short-Circuit Current	14.04	A	Isc

Table 11. STC electrical characteristics.

Parameter	Value	Unit	Abbreviation
Output Power	555	W	Pmax
Output Power Tolerance	0/+5	W	Δpmax
Module Efficiency	21.48	%	Hm
Voltage at Pmax	42.24	V	Vmpp
Current at Pmax	13.14	A	Impp
Open-Circuit Voltage	49.9	V	Voc
Short-Circuit Current	14.04	A	Isc

Table 12. Coefficient parameters of the 555 W panel.

Temperature Coefficient		Value
α	Temperature Coefficient Voc (%/°C)	−0.29
β	Temperature Coefficient Isc (%/°C)	0.048
γ	Temperature Coefficient Pmax (%/°C)	−0.36

Table 12. Coefficient parameters of the 555 W panel.

Temperature Coefficient		Value
α	Temperature Coefficient Voc (%/°C)	−0.29
β	Temperature Coefficient Isc (%/°C)	0.048
γ	Temperature Coefficient Pmax (%/°C)	−0.36

After obtaining the corrected parameters, it is necessary to cover the maximum 746 kW under the least favorable conditions (lower irradiation and a loss factor of 0.7). The application of CanadianSolar 125 kW three-phase inverters is proposed, requiring six units of this technology.

Table 13. Inverter feature.

Parameter	Value
Nominal Output Power kW	125
Max PV Power kW	187.5
Max DC Voltage V	1500
Stard Voltage V	860
PV Voltage Renge V	860–1450
Max Input Current A	300
# Of MPP Trankers	1

shows the key details of the technical data.

Using the data from the inverter table characteristics and the temperature-related voltage corrections, the connections of the panels in strings (in series) and chains (in parallel) are approximated. Using Equations (17) to (21). In Table 14, the calculated results are observed.

$$string={\displaystyle \frac{Vrangemax}{V_{oc}min}}$$

(17)

$$P_{strmin}=String\ast P_{Tc}min$$

(18)

$$P_{strmax}=String\ast P_{Tc}max$$

(19)

$$\mathit{Chain\_min}={\displaystyle \frac{P_{nominal}}{P_{strmin}}}$$

(20)

$$\mathit{Chain\_max}={\displaystyle \frac{P_{nominal}}{P_{strmax}}}$$

(21)

• With a $V_{oc}min=46.43 \mathrm{V}$ to reach the maximum voltage range of 1450 V.
• With a $V_{oc}min=46.43 \mathrm{V}$ to reach the 860 V range required.
• With a $V_{oc}max=50.73 \mathrm{V}$ to reach a range of 1450 V.
• With a $V_{oc}max=50.73 \mathrm{V}$ to reach the 860 V range.

Table 14. Estimated data with temperature-corrected parameters.

Parameter	Value
Min. Panels On String	19
Max. Panels On String	29
Voltage Min. (V)	860
Voltage Mav x. (V)	1450
Chain Max.	7–13
Imax (A)	1198

Table 14. Estimated data with temperature-corrected parameters.

Parameter	Value
Min. Panels On String	19
Max. Panels On String	29
Voltage Min. (V)	860
Voltage Mav x. (V)	1450
Chain Max.	7–13
Imax (A)	1198

Figure 14 shows the general diagram and

Table 15. System summary.

Concept	Valor
Num of modules	1368
Module area (m2)	3534
Num of inversors	6
Nominal PV power (kWp)	759
Nominal AC power (kWCA)	750
Power nominal (Pnom)	1.012

summarizes the calculations used to size the photovoltaic system.

3. Results

In order to validate the theoretical sizing obtained and evaluate the actual performance of the photovoltaic system before integrating it into the hybrid model, PVsyst 7.4 software was used as a complementary analysis tool. The results obtained in PVsyst provided a reliable reference for the available energy, expected losses, and seasonal performance of the photovoltaic field, ensuring the technical consistency of the previous sizing. The proposed configuration was reproduced on this platform using 555 Wp modules and 125 kWac three-phase inverters, entering the relevant data in the different sections required by the program, such as generator orientation, system type, detailed losses, self-consumption profile, and optional parameters for shading, module design, and economic evaluation. The simulation made it possible to verify the consistency between the installed capacity and the minimum monthly irradiation of the site, as well as to estimate the effective production under real operating conditions using the established performance factor (PR). Based on the results obtained in PVsyst for the proposed configuration of the photovoltaic system, a set of graphical and numerical analyses was generated to validate the design criteria and characterize the expected performance of the solar field. The corresponding figures show, first, the orientation of the selected inclined plane (20° inclination and 0° azimuth), which guarantees optimal capture for the site’s annual irradiation, showing a transposition factor (TF) of 1.06 and an annual global irradiance on the collector plane of 2063 kWh/m². Subsequently, the PV subsystem design module allows the final configuration of the array to be observed: 555 Wp SOLAREVER modules, operating in conjunction with six 125 kWac SMA inverters, reaching a total nominal power of 759 kWp and adequate electrical dimensioning (Vmpp and Voc voltages within the inverter’s operating ranges). Likewise, the “User Requirements” tool allows the estimated PV generation to be compared with the campus’s monthly consumption, highlighting the marked seasonal variability with peaks exceeding 58,000 kWh in months with higher loads. Finally, the shading analyses confirm favorable conditions for the installation: the diffuse shading factor is low (0.046) and the average albedo is 0.890, while the solar diagram shows that the module plane remains free of obstructions during the hours of highest solar resource throughout the year.

Together, these simulations validate the correct orientation, safe electrical sizing, consistency between demand and generation, and viability of the PV field for subsequent integration into the hybrid model within HOMER Pro, providing a solid technical basis for the energy and economic evaluation of the proposed photovoltaic system. This allows verification that the data calculated and delivered by PVsyst are similar between Table 15 and those shown in Figure 15.

3.1. Estimated Production Costs for Peak Rates

Table 16. Fuel costs per tanker and discount based on purchase percentages.

Pipe Capacity	Retail Price (≈$25.60/L)	Discount Assumption	Estimated Discounted Price
5000 L	$128,000 MXN	5%	$121,600 MXN
		10%	$115,200 MXN
		15%	$108,800 MXN

presents the estimated cost of purchasing a 5000 L tanker of diesel fuel, considering the average retail price of MXN 25.60/L and possible discounts for institutional purchases.

The base cost is approximately $128,000 MXN, while discounts of 5%, 10%, and 15% reduce this amount to $121,600, $115,200, and $108,800 MXN, respectively. Given that the estimated annual consumption is 4121.95 L, this corresponds to 80.3% of a tanker truck, leaving a margin for operational contingencies. The comparative analysis between capacity costs and peak rates shows that the price of fuel remains competitive with the charges applied by the grid, reinforcing the advisability of using the generator during peak hours (see Figure 16). There is also a significant decrease in demand in April, attributed to the holiday period, so it is recommended that the generator be kept out of operation during that month. In economic terms, even under the most conservative scenario, partially replacing grid consumption with diesel generation would reduce annual expenditure by approximately MXN 115,849, equivalent to 51.2% of the total annual cost (MXN 226,230.97).

• Savings with a 5% discount: $126,359 MXN (55.9% of the cost).
• Savings with a 10% discount: $130,580 MXN (57.7% of the cost).
• Savings with a 15% discount: $137,771 MXN (60.8% of the expense).

The images in Figure 17 show the hourly performance of the diesel generator within the hybrid system during a representative day of the year, highlighting the dispatch strategy implemented to operate exclusively during peak hours defined by the hourly rate. The orange curve represents the power generated by the HYUNDAI HYEG36 kW equipment, showing that its operation is concentrated during the hours of highest electricity cost, while it remains out-of-service during the rest of the day thanks to the contribution of the photovoltaic system and the energy supplied from the grid. This optimized management significantly reduces energy costs by shifting the purchase of electricity to the most expensive tariff period. Likewise, it can be seen that the power supplied by the generator dynamically adjusts to the load profile, reaching its highest level during the afternoon peak and maintaining the minimum of 30% established to avoid inefficient operation. In contrast, during the rest of the day, the electricity grid and solar generation (blue and yellow curves) comfortably cover demand, demonstrating adequate coordination between sources and diesel operation restricted only to times when its use is economically advantageous. This behavior confirms that the generator acts as a strategic resource to mitigate the costs associated with peak hours, while ensuring operational continuity and extending its useful life by avoiding unnecessary periods of operation.

3.2. Estimation of Costs and Production for the Intermediate Tariff

The data presented in

Table 17. Energy production with minimum and actual HSP and fatuta consumption (net metering).

Month	HSP	Production with Min. HSP (3.8) kWh/Month	Production with Real HSP kWh/Month	Consumption During Off-Peak Hours Intermediate kW/Month	Net Metering HSP/Month kWh/Month	Net Metering HSP (3.8) kWh/Month
Jan	4.25	61,515.2	68,845.2	21,169	47,676	40,346.2
Feb	5.04	55,562.1	73,625.6	27,625	46,001	27,937.1
Mar	5.84	61,515.2	94,553.7	43,583	50,971	17,932.2
Apr	6.49	59,530.8	101,733.4	41,285	60,448	18,245.8
May	6.51	61,515.2	105,318.8	53,234	52,085	8281.2
Jun	5.37	59,530.8	84,193.8	36,739	47,455	22,791.8
Jul	5.37	61,515.2	86,869.1	19,299	67,570	42,216.2
Aug	5.41	61,515.2	87,497.2	37,092	50,405	24,423.2
Sep	5.26	59,530.8	82,461.1	49,970	32,491	9560.8
Oct	4.86	61,515.2	78,632.6	52,911	25,722	8604.2
Nov	4.34	59,530.8	68,034.3	37,103	30,931	22,427.8
Dec	3.86	61,515.2	62,543.1	21,940	40,603	39,575.2

of Annex E.1 allows for an assessment of the degree of correspondence between the energy generated by a large-scale photovoltaic (PV) system and the monthly consumption recorded in the intermediate period of a medium-voltage tariff system with a time-of-use structure. The analysis was carried out considering two production scenarios: an oversized scenario, calculated based on the minimum value of peak sun hours (HSP) for the site (3.8 h), and a representative scenario based on the specific HSP for each month, derived from the actual available solar resource. Both approaches allow the technical potential of the system to be estimated, both to cover its own consumption and to determine the energy surplus that can be fed into the grid under a net-metering scheme.

Figure 18 presents a graphical comparison between the estimated monthly production in both scenarios, production with minimum HSP (blue line) and production with average monthly HSP (green line), in relation to the electricity demand recorded in the intermediate period (red line). The results show that photovoltaic generation consistently exceeds the facility’s monthly consumption, both under a conservative scenario and under one adjusted to actual solar resources, although with variations in the magnitude of the surplus.

During the months with the highest solar radiation, particularly between March and June, the scenario based on actual HSP reaches production values exceeding 100,000 kWh per month, which shows a high potential for surplus generation that could be used for grid injection or storage. In contrast, the estimated production with the minimum HSP remains between 55,000 and 65,000 kWh per month throughout the year, reflecting the system’s ability to ensure a stable level of generation even under unfavorable solar conditions. On the other hand, the educational facility’s electricity consumption shows significant variations, with peaks during the months of May and September (between 50,000 and 60,000 kWh) and notable decreases in July and December, associated with periods of lower institutional activity. The results show that in both scenarios photovoltaic generation greatly exceeds demand during intermediate hours. For example, during the months of maximum irradiance, such as March, April, and May, production under actual HSP ranges between 94,553 and 105,318 kWh, while the consumption recorded for those same months ranges between 41,285 and 53,234 kWh. This difference generates significant energy surpluses, in the order of 50,000 kWh per month.

The constant availability of surpluses not only ensures coverage of consumption in the interim period, but also has the potential to partially offset consumption corresponding to other times or months with lower irradiation, depending on the compensation mechanism allowed by current regulations. These results indicate that the PV system analyzed is not only technically feasible, but also offers the prospect of a substantial reduction in the total cost of electricity billing. Likewise, the high accumulation of surplus energy would contribute to reducing the operation of the diesel generator and, depending on the final configuration of the hybrid system, would reduce the energy purchased from the grid during peak hours.

In addition to the previous analysis based on peak sunshine hours and the estimated monthly production of the photovoltaic system, it is essential to integrate the results provided by the HOMER Pro simulation, which allow for a more accurate observation of the hourly and annual performance of the PV plant in relation to the campus’s electricity consumption and its interaction with the grid. The graph (Figure 19) generated by the software, corresponding to the output power of the photovoltaic system, shows in detail the distribution of solar generation over the 8760 h of the year, highlighting the actual irradiation patterns using a heat map that distinguishes the instantaneous power levels delivered by the PV system. This representation shows that the plant, with a nominal capacity of 423 kW and an estimated annual production of 659,764 kWh, reaches maximum peaks of up to 416 kW during the months of highest irradiation, while on average it supplies around 1808 kWh/d daily. This behavior coincides with the periods of highest production identified in the monthly HSP assessment, confirming the system’s capacity to amply cover demand during off-peak hours and even generate significant surpluses that can be exported to the grid under net-metering schemes. Likewise, the HOMER results show a solar penetration of 130%, reflecting the high contribution of photovoltaic energy in covering annual demand and its relevance in displacing energy consumption from the grid, reducing diesel generator operation and charges associated with peak and base hours. Overall, this simulation quantitatively supports the viability of the oversized PV system and demonstrates the consistency between theoretical calculations, available solar resources, and actual projected operation within the PV–diesel–grid hybrid system.

4. Discussion

To carry out the evaluation of the energy performance of the photovoltaic system at different levels of renewable penetration (100%, 70%, 50%, and 30%), the installed capacity of 746.05 kWp was established as the starting point, corresponding to the maximum solar integration scenario considered in this research. This value was used as a reference to derive the equivalent configurations of 522 kWp, 372 kWp, and 224 kWp through a proportional reduction in the total power.

These configurations make it possible to analyze how the technical, economic, and environmental performance of the hybrid system varies as a function of the installed photovoltaic capacity, while keeping constant the design criteria, module characteristics, system losses, and operational assumptions used in sizing the base scenario.

Figure 20 presents a detailed comparison of the energy performance at the different levels of renewable penetration, illustrating the relationship between installed capacity, irradiance variability, and the energy utilization potential of the proposed hybrid system.

The results obtained from the economic simulation of the hybrid PV–grid–diesel system allow for a comprehensive evaluation of the financial performance of the proposed design. The model incorporates photovoltaic modules from Canadian Solar (261 kW) and SolarEver (423 kW), a Hyundai HYEG36KW diesel generator, and interconnection to the electrical grid under the utility’s tariff scheme.

The economic analysis reveals that the system has a net present cost (NPC) of $6,960,603.00 and a levelized cost of energy (LCOE) equivalent to $0.7/kWh, values that confirm that, despite the magnitude of the initial investment, the hybrid solution remains competitive compared to conventional supply schemes, especially in tariff contexts with high peak-hour costs.

Additionally, the annual operating cost amounts to $311,912.80, reflecting expenses associated with residual energy consumption, diesel generator operation, and system maintenance. See Figure 21.

The image presented in Figure 22 corresponds to the comprehensive summary of electricity production and consumption of the hybrid photovoltaic system with a diesel generator interconnected with the grid. This result evaluates the overall energy performance of the proposed topology and quantifies the relative contribution of each component during a full year of operation.

In the left block, it is observed that the photovoltaic system is responsible for the largest share of energy production, reaching 659,764 kWh/year (80.7%), confirming its central role as the primary renewable source within the system. The HYEG36KW diesel generator contributes only 17,188 kWh/year (2.1%), reflecting its limited and strategic use (peak hours), mainly during periods of low irradiation or at times of high-capacity demand (excessive cost). The remaining 17.2% comes from purchases from the electrical grid, which ensures supply continuity when neither the photovoltaic system nor the generator can meet the demand and cover the base load.

In the central block, related to consumption, it is shown that the primary AC load absorbs 504,404 kWh/year (67%), while the system exports 248,811 kWh/year (33%) to the grid as surplus photovoltaic energy. This demonstrates a robust performance of the renewable system, capable not only of meeting demand but also of generating significant contributions to the grid under net-metering schemes.

Additionally, minor values of unmet load (1924 kWh/year) and capacity shortage (2407 kWh/year) are identified, both below 0.5%, indicating a reliable and properly sized system. The right block complements the analysis by highlighting a renewable fraction of 79%, demonstrating the high participation of the panels, and a maximum renewable penetration of 159%, associated with periods of higher solar overproduction.

The lower graph, corresponding to monthly electricity production, shows the combined contribution of the photovoltaic system, the diesel generator, and energy from the grid.

Analysis of the diesel generator’s fuel consumption makes it possible to characterize its operational behavior throughout the year. At first, the monthly analysis (upper-left graph) shows that the generator’s hourly consumption presents significant variations between months, with values typically ranging between 2 and 10 L/h, depending on the electrical demand supplied and the load level required during critical operating periods. The highest average consumptions are recorded in February, March, April, and November, which coincide with intervals in which the generator is dispatched more frequently to contain costs associated with temporary increases in electrical demand.

The annual heat map (upper-right graph) provides a detailed visualization of the hourly distribution of consumption throughout the 365 days of the year. It is observed that the generator’s operation is predominantly concentrated within a well-defined time range, corresponding to periods of higher demand on the electrical system. The intensity of consumption, although variable, remains within expected ranges according to the equipment’s efficiency curve. Yellow and red tones indicate conditions of higher electrical and thermal load, while green and blue tones represent partial-load or transient operating states.

Meanwhile, the daily time series (lower graph) complements the analysis by explicitly showing each generator operating event throughout the year. In this graph, recurring peaks are identified during months of higher energy demand, confirming that the generator does not operate continuously, but under a controlled and selective dispatch strategy. This strategy makes it possible to significantly limit the total operating time of the equipment, optimizing its use and reducing both fuel consumption and mechanical wear.

Total annual diesel consumption amounts to 4704 L, which is equivalent to an average of 12.9 L/day and 0.537 L/h, considering only actual operating hours. It should be noted that the software does not distinguish between the two weekend days, affecting the actual figure of 4014.88 L (peak rates are not applied on Sundays and holidays). These values are consistent with the results calculated from the simulation model and validate the consistency of the dispatch scheme adopted; see Figure 23.

The annual cash flow chart generated (Figure 24) allows visualization of the evolution of the costs and economic benefits of the hybrid system over the project time horizon. First, a significant initial investment is observed, represented by the bar corresponding to year zero, associated with the capital cost of the system’s main components: the 423 kW photovoltaic field, the 261 kW SMA inverters, and the 36 kW HYUNDAI HY36kW diesel generator. This investment, close to 2.8 million pesos, constitutes the largest expenditure and establishes the starting point of the financial analysis.

Throughout the subsequent years, the lower bars represent recurring operating costs, mainly associated with preventive and corrective maintenance, administrative costs, and a reduced fraction corresponding to the annual fuel consumption of the diesel generator, whose use is limited due to the high contribution of the photovoltaic system. Although negative, these bars show relatively constant and low-magnitude values, confirming the low-operating-cost nature of renewable-based systems.

In year 15, a significant, extraordinary expense is observed corresponding to the scheduled replacement of critical system components, identified by the deeper orange bar. This replacement is part of the projected service life of the inverters or auxiliary equipment and is consistent with typical replacement cycles established by manufacturers. Despite this temporary increase in costs, the system’s operational stability is not compromised.

At the end of the project’s useful life, in year 25, a positive bar appears representing the salvage value of components that still retain residual value. The economic indicators provided by HOMER Pro, detailed in the upper right section of the panel, show a net present cost (NPC) of approximately $6.96 million, a levelized cost of energy (LCOE) of $0.707/kWh, and an annual operating cost close to $319,000 MNM.

The obtained LCOE is notably lower than the average cost of the medium-voltage industrial tariff in Mexico, which demonstrates the competitiveness of the proposed system. Likewise, comparison with the base scenario (without PV system) shows a cost reduction greater than 50%, validating the economic viability of the hybrid system. In other words, the cash flow behavior shows that, despite the high initial investment, the system achieves solid financial recovery, with sufficient income and savings to offset operating and replacement costs throughout its useful life.

The reduction in NPC compared to the base case, together with an LCOE lower than the electricity purchase price from CFE, indicates that the investment is financially profitable and highly favorable. In energy terms, the high renewable fraction (≈79%) substantially reduces dependence on the diesel generator and the electrical grid, contributing to the system’s economic stability under tariff volatility scenarios.

In summary, the cash flow confirms that the project is not only technically and energetically viable, but also financially profitable, providing sustained economic benefits over the 25 years of evaluated operation, view Figure 24.

The economic results of each scenario allow evaluation of the financial performance of the photovoltaic system under different renewable generation capacities (100%, 70%, 50%, and 30%), as well as the case optimized by HOMER Pro, in comparison with the base system connected only to the grid, which generates an annual cost of $1,313,579.7 MXN.

Table 18. Comparative matrix of photovoltaic system scenarios.

PV Scenario	Behavior Under Minimum HSP (3.8 h)	Behavior Under Real Monthly HSP	Shaded Area (Surplus vs. Deficit)	Intermediate Demand Coverage	Grid Dependency	Global Technical Interpretation (Doctoral Criterion)
100% (746 kWp)	PV generation exceeds consumption in almost all months.	Generation increases significantly during high-irradiance months (Mar–Sep).	Large positive shaded area throughout the year → significant surpluses.	Full coverage with monthly surpluses.	Minimal—nearly independent of the grid.	Oversized system for the intermediate tariff period; maximizes net-metering credits and reduces annual operating costs.
70% (522 kWp)	Generation approaches consumption during low irradiance months; moderate surpluses in high months.	Generation exceeds demand in several months (Mar–Jun and Aug).	Positive shaded area during high-irradiance months; reduced during low months.	High coverage; only a few months require additional energy.	Low, with occasional grid support.	Represents a robust and well-balanced configuration between investment, generation, and economic return.
50% (372 kWp)	Generation remains below consumption for most months.	Surpluses occur only during very-high-irradiance months.	Limited shaded area; deficit during most of the year.	Partial coverage; higher dependence on the grid.	Moderate, especially during winter months.	Intermediate configuration: reduces energy costs and emissions but does not achieve energy self-sufficiency under intermediate tariff conditions.
30% (224 kWp)	Generation is consistently lower than intermediate monthly demand.	Even under real HSP, generation does not exceed demand in any month.	No surpluses; shaded area always negative (deficit).	Low coverage; only partial contribution.	High, requiring grid energy for most demand.	Suitable for partial peak reduction and cost savings, but unable to generate surpluses for net metering.

presents a comparative analysis of different levels of photovoltaic system sizing, evaluating their performance relative to demand during the intermediate tariff period. It can be observed that, as the installed capacity increases, demand coverage and surplus generation increase, significantly reducing dependence on the power grid. The 100% scenario indicates oversizing, with constant surpluses throughout the year, while the 70% scenario represents an optimal balance between generation, consumption, and investment. On the other hand, the 50% and 30% sizing scenarios show deficits in most months, which increases dependence on the grid. Overall, the table demonstrates the direct relationship between the sizing of the photovoltaic system, the level of energy self-sufficiency, and its technical–economic impact under conditions of variable irradiance.

Table 19. Scenarios and economic analysis of the VPN and IRR.

Scenario	Initial Investment	Estimated Payback	VPN (MXN)	IRR (%)	Comment
100%	High $9,083,442	Year 6	$5,229,639.72	15%	Profitable, but lower economic efficiency than 70%
70%	Moderate $6,833,442	Year 5	$6,107,403.56	18%	Best PV scenario, optimal cost–benefit balance
50%	Medium-low $5,470,942	Year 7	$4,986,166.05	18%	Profitable, but with limited benefits
30%	Lowest $4,098,442	Year 10	$3,096,827.08	17%	Marginal profitability; insufficient PV capacity
HOMER Pro	Optimized $6,643 442	Year 4–5	$6,283,329.49	19%	Best overall performance; optimized configuration

presents the amounts for each scenario and the economic analysis of the VPN and IRR.

The VPN represents the present value of the net benefits generated during the system’s useful life, discounting the initial investment. A positive VPN, as occurs in all scenarios, indicates that the project is profitable. The HOMER Pro scenario presents the highest VPN ($6,283,329.49), followed by the 70% scenario ($6,107,403.56), confirming that these two represent the most economically attractive configurations. The 100% scenario obtains an VPN of $5,229,639.72, while the scenarios with a lower installed capacity (50% and 30%) present smaller values due to lower savings relative to the initial investment.

On the other hand, the IRR expresses the project’s average annual profitability. The HOMER Pro scenario reaches the highest IRR (19%), demonstrating superior financial efficiency. The 70% and 50% scenarios have IRRs of 18%, while the 100% scenario presents 15% and the 30% scenario reaches 17%. This behavior reveals that, although PV capacity directly influences energy savings, system optimization—such as that incorporated in HOMER Pro—significantly improves profitability without the need to oversize the photovoltaic installation.

4.1. Sensitivity Variables for VPN and IRR in the 100% Scenario

The sensitivity

Table 20. VPN sensitivity with the variables of cost reduction and revenue increase (100%).

		Cost Reduction
VPN	5,229,639.71	0%	2%	5%	8%	10%
Income increase	0%	5,229,639.71	5,256,694	5,297,276	5,337,857	5,364,912
	2%	5,529,499	5,556,553	5,597,135	5,637,717	5,664,771
	5%	5,979,287	6,006,342	6,046,924	6,087,505	6,114,560
	8%	6,429,076	6,456,131	6,496,712	6,537,294	6,564,348
	10%	6,728,935	6,755,990	6,796,571	6,837,153	6,864,207

shows how the project’s Net Present Value (VPN) varies under two types of changes: a reduction in investment and operating costs (top column) and an increase in project-derived revenues (left row). The base case or original scenario corresponds to an VPN of $5,229,639.71, from which variations of ±0–10% in each variable are evaluated.

When analyzing the percentage variations, it is observed that both variables generate increases in VPN as revenues rise or costs decrease, which is expected in financial models. However, the magnitude of the change makes it possible to determine which of the two has the greater impact on project profitability—that is, which variable is more sensitive.

Revenue increases show much stronger rises in VPN than those observed from an equivalent reduction in costs. For example, a 10% increase in revenues raises the VPN to $6,728,935, whereas a 10% reduction in costs increases it only to $5,364,912. The difference between both effects exceeds one million pesos, clearly showing that the project responds more elastically to variations in revenues than to cost reductions. This trend remains consistent across all analyzed ranges. With a 5% increase in revenues, the VPN rises to $5,979,287, while a 5% reduction in costs brings it to $5,297,276, again showing a much smaller impact from cost reduction. Even when both variables are combined, the most decisive change continues to be associated with revenue increases. In other words, for each fixed level of cost reduction, the marginal increase in VPN produced by raising revenues is greater than that produced by further reducing costs by the same percentage value.

The overall behavior of Table 19 shows that the IRR increases in all cases as revenues rise or costs decrease. However, the magnitude of the change makes it possible to identify which of these two variables has a more significant impact on the project’s profitability.

When comparing percentage variations, it is observed that revenue increases produce larger rises in IRR, while cost reductions generate more moderate changes. For example, with a 10% increase in revenues, the IRR rises to 16.5%, representing an increase of 1.8 percentage points compared to the base value. In contrast, a 10% reduction in costs increases the IRR to only 14.9%, which represents a change of just 0.2 percentage points.

Even when both effects are combined, the dominant impact continues to be associated with revenue increases. This pattern is repeated throughout the table: for each level of cost reduction, the marginal increase in IRR produced by raising revenues is significantly greater than the increase generated by an equivalent percentage change in cost reduction.

The VPN and IRR are considerably more sensitive to increases in revenues than to equivalent reductions in costs. This suggests that strategies aimed at increasing revenues, such as maximizing self-consumption, reducing technical losses, improving operational efficiency, or negotiating more favorable tariffs, provide a much more significant impact on the overall profitability of the system than those focused solely on reducing investment or maintenance costs.

4.2. Sensitivity Variables for VPN and IRR in the HOMER Pro Scenario

Table 21. IRR sensitivity with the variables of cost reduction and revenue increase (100%).

		Cost Reduction
IRR	14.7%	0%	2%	5%	8%	10%
Income Increase	0%	14.7%	14.7%	14.8%	14.8%	14.9%
	2%	15.1%	15.1%	15.1%	15.2%	15.2%
	5%	15.6%	15.6%	15.7%	15.7%	15.8%
	8%	16.1%	16.2%	16.2%	16.3%	16.3%
	10%	16.5%	16.5%	16.6%	16.6%	16.7%

and

Table 22. VPN sensitivity with the variables of cost reduction and revenue increase (HOMER Pro).

		Cost Reduction
VPN	6,283,329.48	0%	2%	5%	8%	10%
Income Increase	0%	6,283,329.48	6,283,329	6,283,329	6,283,329	6,283,329
	2%	6,559,179	6,559,179	6,559,179	6,559,179	6,559,179
	5%	6,972,952	6,972,952	6,972,952	6,972,952	6,972,952
	8%	7,386,726	7,386,726	7,386,726	7,386,726	7,386,726
	10%	7,662,575	7,662,575	7,662,575	7,662,575	7,662,575

present the sensitivity data for the HOMER Pro optimization scenario. From these data, and in comparison, with the 100% scenario, it can be concluded that revenue increases are the most influential variable in both scenarios, for both VPN and IRR. Cost reductions have a marginal impact, since the project’s financial structure concentrates benefits in operational cash flows rather than in the initial investment.

The scenario optimized by HOMER Pro shows greater robustness, as its VPN and IRR present lower variability when facing changes in costs and revenues. In economic terms, the percentage-based scenario is more vulnerable to changes in revenues and shows a more sensitive VPN, whereas HOMER Pro distributes cash flows more efficiently and reduces project volatility.

The IRR in the HOMER Pro scenario is higher in all cases, demonstrating that energy optimization has a stronger positive impact than simply increasing photovoltaic capacity (view

Table 23. IRR sensitivity with the variables of cost reduction revenue increase (Homer Pro).

		Cost Reduction
IRR	14.7%	0%	2%	5%	8%	10%
Income Increase	0%	14.7%	14.7%	14.8%	14.8%	14.9%
	2%	15.1%	15.1%	15.1%	15.2%	15.2%
	5%	15.6%	15.6%	15.7%	15.7%	15.8%
	8%	16.1%	16.2%	16.2%	16.3%	16.3%
	10%	16.5%	16.5%	16.6%	16.6%	16.7%

).

5. Conclusions

This study demonstrates that the integration of a hybrid photovoltaic–grid–diesel system, optimized using the HOMER Pro platform, is a technically and economically viable alternative for reducing energy costs in facilities subject to complex tariff schemes and charges associated with electricity demand. Detailed analysis of the load profile, together with the characterization of energy costs and comparative simulation of multiple photovoltaic penetration scenarios, showed that the profitability of the system does not depend exclusively on the size of the installation, but on the dynamic interaction between renewable generation, demand costs, and the generator’s operating strategy during critical periods.

The results indicate that the system optimized using HOMER Pro offers the best overall economic performance, reflected in a higher Net Present Value and Internal Rate of Return compared to scenarios defined solely by fixed percentages of photovoltaic capacity, as well as a significantly shorter payback period. Likewise, the sensitivity analysis confirmed that the project’s profitability is strongly influenced by the increase in energy revenues or savings derived from self-consumption, while the reduction in initial investment costs has a relatively marginal impact. This behavior is explained by the temporary nature of cash flows, where recurring operating profits carry greater weight than the capital investments made at the start of the project.

Overall, the results validate the methodological relevance of using energy optimization tools for the sizing of hybrid systems, while highlighting the importance of strategies aimed at maximizing self-consumption and minimizing exposure to electricity costs during periods of peak system demand. The study concludes that the proposed system not only efficiently meets energy demand, but also generates sustained economic benefits throughout the project’s lifetime, even considering scenarios of technological replacement halfway through the operating cycle. In this sense, the integration of renewable sources, combined with intelligent energy dispatch design, is consolidating itself as a robust path toward more efficient, sustainable, and financially resilient energy infrastructures.